摘要
将太阳系多目标探测的轨道优化设计问题转换成非线性规划问题,建立了轨道优化模型。针对非线性规划问题解的多峰性,设计了一种融合改进的网格搜索算法和差分进化算法的组合优化算法。利用改进的网格搜索算法以适当的步长寻找理想的发射窗口和各阶段转移时间,产生差分进化的初始群体,进而使用差分进化算法搜索初始群体附近的子空间,通过全局范围内的比较得到较理想的结果。最后以2018～2020年太阳系多目标探测为例,面向土星环绕探测任务完成了飞行中途探测太阳系多颗大行星的轨道优化设计。数值仿真结果表明上述算法对太阳系多目标探测轨道优化设计具有较好的通用性和应用参考价值。
The trajectory optimization design problem of multi-objective detection in solar system is transformed as a nonlinear programming problem, and the trajectory optimization model is derived in this paper. A combinatorial optimization algorithm consists of improved grid search algorithm and differential evolution algorithm, which is proposed to solve many local minima of the nonlinear programming problem. The improved grid search algorithm is employed with appropriate steps, the ideal launch window and transfer time in each phase are obtained, and initial group for differential evolution is produced. The differential evolution algorithm is used to search the vicinity of the initial population. Through the comparison in the global space, the ideal results are obtained. Taking the multi-objective detection from 2018 to 2020 as an exam- ple, the results of the trajectory detection of several giant planets are gained in solar system by detecting around Saturn. The numerical simulation results show that this algorithm has good versatility and application reference value for deep space exploration trajectory optimization design.
出处
《航天控制》
CSCD
北大核心
2013年第4期39-45,共7页
Aerospace Control
基金
国家高技术研究发展计划(863计划)资助(2012AA121602)
国家自然科学基金资助(11078001)
中央高校基本科研业务费专项资金资助(NS2012132)
关键词
多目标探测
轨道优化设计
非线性规划
差分进化
Multi-objective
Trajectory optimization design
Nonlinear programming problem
Differential evolution
